W
HeyGen Voice
HeyGen Voice, a text-to-speech model from WaveSpeed.
speechModelOverride: { model: "heygen-voice" }
HeyGen·served via
WaveSpeed·Released October 2026Pricing
- List price
- $0.04/request
- Per generation
- $0.00010
Specifications
- Context window
- 5,000 tokens
- Released
- October 2026
- Model string
- heygen/voice
Parameters
Model options, passed in
speechModelOverride.config.| Parameter | Type | Default | Range / options |
|---|---|---|---|
voiceA HeyGen voice. Ignored when a reference recording is given. | Select | Jonah | JonahNadineTabithaSorenElowenVincent+18 |
reference_audioA recording of one speaker to clone in place of a HeyGen voice. Only its first 3 minutes are used. | Audio URL | — | — |
languageThe language of the text. Detected from the text when unset. | Select | — | arbebgcazhhr+33 |
expressiveness_boostHow expressive the delivery is, from 0 to 1. | Number | 1 | 0–1, step 0.1 |
Voices · 24
Use this model
await mindstudio.textToSpeech({ text: "...", speechModelOverride: { model: "heygen-voice", config: { voice: "Jonah", expressiveness_boost: 1, }, }, });